Press n or j to go to the next uncovered block, b, p or k for the previous block.
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 | 17x 17x 17x 17x 17x 17x 17x 17x 17x 8x 8x 3x 2x 1x 1586x 1586x 1586x 18549x 18549x 1x 18548x 18548x 18548x 1585x 13x 13x 1x 12x 12x 1572x 1555x 1555x 1x 1554x 1583x 1563x 1563x 1563x 32195x 32195x 1x 32194x 32194x 1562x 41x 41x 1x 40x 1561x 5x 5x 7x 7x 7x 7x 7x 5x 6636x 5x 3144x 3144x 3144x 3144x 563x 563x 563x 563x 7191x 7191x 6652x 8x 8x 3x 3x 3x 3x 1x 2x 2x 2x 2x 2x 2x 2x 6636x 1x 6636x 7191x 7191x 1586x 1586x 1586x 1563x 1563x 1563x 3482x 3482x 3482x 5x 5x 5x 6630x 6630x 7191x 554x 1x 553x | import { ALPHANUMERIC_CHARSET, charCountBits } from "@usefy/qr-code/headless";
import { QRDecodeError } from "../errors";
import type { QRScanSegment, QRSegmentMode, StructuredAppendInfo } from "../types";
import { BitReader } from "./bitreader";
import { decodeSegmentText, decodeWithLabel } from "./charset";
/**
* The QR bit stream parser (ISO/IEC 18004 §7.4) — the inverse of
* `@usefy/qr-code`'s `bitstream.ts` and `segment.ts`, plus the modes that
* package deliberately does not *write*.
*
* A generator only has to emit modes it chose; a scanner meets whatever the
* rest of the world produced. So Kanji, Structured Append and both FNC1
* positions are all handled here even though the counterpart encoder never
* emits them (SPEC decision #7).
*/
/** Mode indicators (ISO Table 2). */
const MODE_TERMINATOR = 0b0000;
const MODE_NUMERIC = 0b0001;
const MODE_ALPHANUMERIC = 0b0010;
const MODE_BYTE = 0b0100;
const MODE_KANJI = 0b1000;
const MODE_ECI = 0b0111;
const MODE_STRUCTURED_APPEND = 0b0011;
const MODE_FNC1_FIRST = 0b0101;
const MODE_FNC1_SECOND = 0b1001;
export interface ParsedBitStream {
readonly segments: QRScanSegment[];
readonly structuredAppend?: StructuredAppendInfo;
readonly fnc1?: "gs1" | "aim";
readonly fnc1Application?: number;
}
/**
* Read an ECI assignment number, whose width is announced by its own leading
* bits: 1 byte for 0–127, 2 for 128–16383, 3 above that.
*/
function readEci(reader: BitReader): number {
const first = reader.read(8);
if ((first & 0x80) === 0) return first & 0x7f;
if ((first & 0xc0) === 0x80) return ((first & 0x3f) << 8) | reader.read(8);
if ((first & 0xe0) === 0xc0) return ((first & 0x1f) << 16) | reader.read(16);
throw new QRDecodeError("parse", `Malformed ECI designator (leading byte 0x${first.toString(16)}).`);
}
/** Numeric mode: three digits per 10 bits, with 7- and 4-bit tails. */
function readNumeric(reader: BitReader, charCount: number): Uint8Array {
const out = new Uint8Array(charCount);
let index = 0;
while (charCount - index >= 3) {
const triple = reader.read(10);
if (triple >= 1000) {
throw new QRDecodeError("parse", `Numeric triplet ${triple} is out of range (0–999).`);
}
out[index++] = 0x30 + Math.floor(triple / 100);
out[index++] = 0x30 + (Math.floor(triple / 10) % 10);
out[index++] = 0x30 + (triple % 10);
}
if (charCount - index === 2) {
const pair = reader.read(7);
if (pair >= 100) {
throw new QRDecodeError("parse", `Numeric pair ${pair} is out of range (0–99).`);
}
out[index++] = 0x30 + Math.floor(pair / 10);
out[index++] = 0x30 + (pair % 10);
} else if (charCount - index === 1) {
const digit = reader.read(4);
if (digit >= 10) {
throw new QRDecodeError("parse", `Numeric digit ${digit} is out of range (0–9).`);
}
out[index++] = 0x30 + digit;
}
return out;
}
/** Alphanumeric mode: two characters per 11 bits, with a 6-bit tail. */
function readAlphanumeric(reader: BitReader, charCount: number): Uint8Array {
const out = new Uint8Array(charCount);
let index = 0;
while (charCount - index >= 2) {
const pair = reader.read(11);
if (pair >= 45 * 45) {
throw new QRDecodeError("parse", `Alphanumeric pair ${pair} is out of range (0–2024).`);
}
out[index++] = ALPHANUMERIC_CHARSET.charCodeAt(Math.floor(pair / 45));
out[index++] = ALPHANUMERIC_CHARSET.charCodeAt(pair % 45);
}
if (charCount - index === 1) {
const value = reader.read(6);
if (value >= 45) {
throw new QRDecodeError("parse", `Alphanumeric value ${value} is out of range (0–44).`);
}
out[index++] = ALPHANUMERIC_CHARSET.charCodeAt(value);
}
return out;
}
/**
* Kanji mode: 13 bits per character, holding a Shift-JIS double byte offset by
* one of two bases. Reassembled into Shift-JIS bytes and handed to
* `TextDecoder("shift_jis")` — no encoding table ships in this package.
*/
function readKanji(reader: BitReader, charCount: number): Uint8Array {
const out = new Uint8Array(charCount * 2);
for (let i = 0; i < charCount; i++) {
const value = reader.read(13);
let assembled = ((value / 0xc0) << 8) | (value % 0xc0);
assembled += assembled < 0x1f00 ? 0x8140 : 0xc140;
out[i * 2] = (assembled >> 8) & 0xff;
out[i * 2 + 1] = assembled & 0xff;
}
return out;
}
/**
* Character-count field widths (ISO Table 3).
*
* `charCountBits` from `@usefy/qr-code` covers the three modes that encoder
* writes; Kanji has its own, narrower widths and is added here. Getting this
* wrong does not produce a wrong character — it desynchronises the whole
* remaining bit stream — so it is a table, not an approximation.
*/
function countBits(mode: QRSegmentMode, version: number): number {
if (mode !== "kanji") return charCountBits(mode, version);
return version <= 9 ? 8 : version <= 26 ? 10 : 12;
}
/** ASCII/Latin-1 bytes → string, chunked so a 7,089-digit payload cannot blow the stack. */
function latin1(bytes: Readonly<Uint8Array>): string {
let out = "";
for (let i = 0; i < bytes.length; i += 8192) {
out += String.fromCharCode(...bytes.subarray(i, Math.min(i + 8192, bytes.length)));
}
return out;
}
/**
* Parse the corrected data codewords into segments.
*
* Consecutive segments of the same mode and ECI are **not** merged — the
* segmentation is part of what the symbol said, and a consumer inspecting
* `segments` should see what was actually encoded.
*
* @param data - Error-corrected data codewords.
* @param version - Symbol version, which sets the character-count field widths.
* @throws {QRDecodeError} with stage `"parse"` on an unknown mode or a stream
* that ends mid-segment.
*/
export function parseSegments(data: Readonly<Uint8Array>, version: number): ParsedBitStream {
const reader = new BitReader(data);
const segments: QRScanSegment[] = [];
let structuredAppend: StructuredAppendInfo | undefined;
let fnc1: "gs1" | "aim" | undefined;
let fnc1Application: number | undefined;
let eci: number | undefined;
// A symbol whose data section is entirely padding is not a decode failure in
// the mathematical sense, but it is not a QR payload either.
let sawSegment = false;
while (reader.available >= 4) {
const mode = reader.read(4);
if (mode === MODE_TERMINATOR) break;
switch (mode) {
case MODE_ECI:
eci = readEci(reader);
continue;
case MODE_STRUCTURED_APPEND: {
const index = reader.read(4);
const total = reader.read(4) + 1;
const parity = reader.read(8);
if (index >= total) {
throw new QRDecodeError(
"parse",
`Structured Append says symbol ${index + 1} of ${total}, which is impossible.`,
);
}
structuredAppend = { index, total, parity };
continue;
}
case MODE_FNC1_FIRST:
fnc1 = "gs1";
continue;
case MODE_FNC1_SECOND:
fnc1 = "aim";
fnc1Application = reader.read(8);
continue;
case MODE_NUMERIC:
case MODE_ALPHANUMERIC:
case MODE_BYTE:
case MODE_KANJI:
break;
default:
throw new QRDecodeError(
"parse",
`Unknown mode indicator 0b${mode.toString(2).padStart(4, "0")} in the bit stream.`,
);
}
const segmentMode: QRSegmentMode =
mode === MODE_NUMERIC
? "numeric"
: mode === MODE_ALPHANUMERIC
? "alphanumeric"
: mode === MODE_BYTE
? "byte"
: "kanji";
const count = reader.read(countBits(segmentMode, version));
let bytes: Uint8Array;
let text: string | undefined;
switch (segmentMode) {
case "numeric":
bytes = readNumeric(reader, count);
text = latin1(bytes);
break;
case "alphanumeric":
bytes = readAlphanumeric(reader, count);
text = latin1(bytes);
break;
case "byte":
bytes = reader.readBytes(count);
text = decodeSegmentText(bytes, eci);
break;
case "kanji":
bytes = readKanji(reader, count);
text = decodeWithLabel(bytes, "shift_jis");
break;
}
sawSegment = true;
// The ECI is reported only where it actually determined the text. Numeric
// and alphanumeric segments are ASCII by definition, and Kanji is Shift-JIS
// by definition, so tagging those with whatever ECI happened to be in force
// would be noise a consumer has to learn to ignore.
const carriesEci = eci !== undefined && segmentMode === "byte";
segments.push(
carriesEci ? { mode: segmentMode, text, bytes, eci } : { mode: segmentMode, text, bytes },
);
}
if (!sawSegment) {
throw new QRDecodeError("parse", "The symbol carries no data segments.");
}
return { segments, structuredAppend, fnc1, fnc1Application };
}
|